OBJECT PATH IDENTIFICATION FOR NAVIGATING OBJECTS IN SCENE-AWARE DEVICE ENVIRONMENTS

• US 20170132839A1
• Filed: 11/09/2015
• Published: 05/11/2017
• Est. Priority Date: 11/09/2015
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Kohler, Jeffrey, Ambrus, Anthony James
• Granted Patent: US 9,928,648 B2
• Time in Patent Office: Days
• CPC Class Codes: G02B 2027/0138 comprising image capture systems, e.g. camera
• CPC Class Codes: G02B 2027/014 comprising information/image processing systems
• CPC Class Codes: G02B 2027/0141 characterised by the informative content of the display
• CPC Class Codes: G02B 27/0172 characterised by optical features
• CPC Class Codes: G06T 19/003 Navigation within 3D models or images
• CPC Class Codes: G06T 19/006 Mixed reality object pose determination, tracking or camera calibration for mixed reality G06T7/00
• CPC Class Codes: G06T 2210/21 Collision detection, intersection
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes

Abstract

In various embodiments, computerized methods and systems for identifying object paths to navigate objects in scene-aware device environments are provided. An object path identification mechanism supports identifying object paths. In operation, a guide path for navigating an object from the start point to the end point in a scene-aware device environment is identified. A guide path can be predefined or recorded in real time. A visibility check, such as a look-ahead operation, is performed based on the guide path. Based on performing the visibility check, a path segment to advance the object from the start point towards the end point is determined. The path segment can be optionally modified or refined based on several factors. The object is caused to advance along the path segment. Iteratively performing visibility checks and traverse actions moves the object from the start point to the end point. The path segments define the object path.

Claims

1. A computer-implemented method for identifying object paths to navigate objects in scene-aware device environments, the method comprising:

identifying a guide path for navigating an object from a start point to an end point in a scene-aware device environment;

executing a visibility check operation based on the guide path;

determining a path segment to advance the object from the start point towards the end point, wherein the path segment comprises a first point and a second point, the second point identified based on executing the visibility check operation; and

causing the object to advance along the path segment.

2. The method of claim 1, wherein identifying the guide path is based on real time tracking to define a real time guide path from the start point to the end point.

3. The method of claim 2, wherein the real time guide path comprises an anterior path portion, a central path portion, and posterior path portion, wherein the anterior path portion and the posterior path portion are portions of the real time guide path that are not explicitly traversed when generating the real time guide path.

4. The method of claim 1, wherein identifying the guide path is based on receiving a user indicated guide path for the scene-aware environment from the start point to the end point.

5. The method of claim 1, wherein selecting the guide path is based on comparing a set of coordinates of the guide path to a sets of coordinates of the start point and a set of coordinates of the end point in a scene-aware device environment.

6. The method of claim 1, executing a visibility check comprises:

determining that a look-ahead point on the guide path is an un-occluded point from a current position of the object, wherein the look-ahead point is proximate an occluded point on the guide path; and

selecting the look-ahead point as a next point of the object path for defining the path segment, the path segment is defined based on the current position and the look-head point such that the object traverses the path segment.

7. The method of claim 1, further comprising smoothing the path segment based on executing a smoothing operation comprising:

determining a lower bound that represents an un-occluded point that is lowest on the guide path with reference to a current position of the object;

determining an upper bound that represents a subsequent point on the guide path that is un-occluded from the lower bound; and

identifying a selected point between the lower bound and the upper bound, wherein when the selected point is occluded the selected point is designated as a new lower bound and when the selected point is un-occluded the selected point is designated as a new upper bound.

8. The method of claim 1, further comprising relaxing the path segment based on executing a relaxation operation comprising:

selecting an adjustment pivot point on the guide path;

determining a height of a previous point of the guide path;

determining a height of the next point of the guide path;

calculating an adjusted height based on an average of the height of the previous point and the height of the next point;

determining that the adjustment pivot point at the adjusted height, the previous point, and the next point are each un-occluded; and

designating the adjusted height as a new height for the next point.

9. The method of claim 1, further comprising motion adapting the path segment based on executing a motion adaptation operation comprising:

determining a motion feature corresponding to movement associated with the object; and

applying the motion feature, based on attributes of the motion feature, to the path segment such that the object path is adapted to incorporate the motion feature.

10. The method of claim 1, further comprising:

detecting an occlusion in a path segment previously identified as un-occluded; and

communicating an indication that the path segment is occluded to facilitate redefining the path segment as a new path segment that is un-occluded.

11. One or more computer storage media having computer-executable instructions embodied thereon that, when executed, by one or more processors, causes the one or more processors to execute operations for identifying object paths to navigate objects in a scene-aware device environment, the operations comprising:

identifying a guide path for navigating an object from a start point to an end point in a scene-aware device environment, wherein the guide path is selected based on a set of coordinates of the start point and a set of coordinates of the end point;

executing a visibility check operation that determines that a look-ahead point on the guide path is an un-occluded point, the visibility check operation comprising casting in the scene-aware environment based on attributes of the object traversing the path segment;

determining a path segment to advance the object from the start point towards the end point, wherein the path segment comprises a first point and a second point, the second point identified based on executing the visibility check operation; and

causing the object to advance along the path segment.

12. The media of claim 11, wherein casting is executed in executing the visibility check operation based on attributes of an anticipated object utilizing the guide path.

13. The media of claim 11, wherein executing a visibility check comprises:

determining that a look-ahead point on the guide path is an un-occluded point from a current position of the object; and

selecting the look-ahead point as a next point of the object path for defining the path segment, the path segment is defined based on the current position and the look-head point such that the object traverses the path segment.

14. The media of claim 11, further comprising iteratively executing both a smoothing operation to smoothen the path segment and a relaxation operation to relax a height of the path segment.

15. The media of claim 11, wherein causing the object to advance along the path segment further comprises:

casting in the scene-aware environment based on attributes of the object corresponding to movement associated with the object;

applying the motion feature, based on attributes of the motion feature, to the path segment such that the object path is adapted to incorporate the motion feature; and

causing the object to traverse along an adapted path segment based on apply the motion feature.

16. A system for identifying object paths to navigate objects in scene-aware device environments, the system comprising:

a processor and a memory configured for providing computer program instructions to the processor;

a path-navigation component configured to;

identify a guide path for navigating an object from a start point to an end point in a scene-aware device environment;

execute a visibility check operation using the guide path;

determine a path segment to advance the object from the start point towards the end point, wherein the path segment comprises a first point and a second point based on executing the visibility check operation; and

causing the object to advance along the path segment.

17. The system of claim 16, further comprising:

a capture component configured to;

capture a real world environment to identify the guide path is based on real time tracking, wherein real time tracking operates to define a real time guide path from the start point to the end point, wherein the real time guide path comprises an anterior path portion, a central path portion, and posterior path portion, wherein the anterior path portion and the posterior path portion are portions of the real time guide path that are not explicitly traversed when generating the real time guide path.

18. The system of claim 16, further comprising:

the path-navigation component configured to;

select a guide path from a plurality of predefined guide paths for a scene-aware device environment, the plurality of guide paths include one or more guide paths that are manually defined based on a selection of a path within the scene-aware environment.

19. The system of claim 16, further comprising:

the path-navigation component configured to;

execute supplementary operations, wherein supplementary operations comprises one of;

a smoothing operation comprising;

determining a lower bound that represents an un-occluded point that is lowest on the guide path with reference a current position of the object;

determining an upper bound that represents a subsequent point on the guide path that is un-occluded from the lower bound; and

identifying a selected point between the lower bound and the upper bound, wherein when the selected point is occluded the selected point is designated as a new lower bound and when the selected point is un-occluded the selected point is designated as a new upper bound; and

a relaxation operation comprising;

selecting an adjustment pivot point on the guide path;

determining a height of a previous point of the guide path;

determining a height of the next point of the guide path;

calculating an adjusted height based on an average of the height of the previous point and the height of the next point;

determining that the adjustment pivot at the adjusted height, the previous point, and the next point are each un-occluded; and

designating the adjusted height as a new height for the adjustment pivot point.

20. The system of claim 16, further comprising:

the path-navigation component configured to;

detect an occlusion in a path segment previously identified as un-occluded; and

communicate an indication that the path segment is occluded to facilitate redefining the path segment as a new path segment that is un-occluded.